Events & Conferences
Latency from post-quantum cryptography shrinks as data increases
The risk that a quantum computer might break cryptographic standards widely used today has ignited numerous efforts to standardize quantum-resistant algorithms and introduce them into transport encryption protocols like TLS 1.3. The choice of post-quantum algorithm will naturally affect TLS 1.3’s performance. So far, studies of those effects have focused on the “handshake time” required for two parties to establish a quantum-resistant encrypted connection, known as the time to first byte.
Although these studies have been important in quantifying increases in handshake time, they do not provide a full picture of the effect of post-quantum cryptography on real-world TLS 1.3 connections, which often carry sizable amounts of data. At the 2024 Workshop on Measurements, Attacks, and Defenses for the Web (MADweb), we presented a paper advocating time to last byte (TTLB) as a metric for assessing the total impact of data-heavy, quantum-resistant algorithms such as ML-KEM and ML-DSA on real-world TLS 1.3 connections. Our paper shows that the new algorithms will have a much lower net effect on connections that transfer sizable amounts of data than they do on the TLS 1.3 handshake itself.
Post-quantum cryptography
TLS 1.3, the latest version of the transport layer security protocol, is used to negotiate and establish secure channels that encrypt and authenticate data passing between a client and a server. TLS 1.3 is used in numerous Web applications, including e-banking and streaming media.
Asymmetric cryptographic algorithms, such as the one used in TLS 1.3, depend for their security on the difficulty of the discrete-logarithm or integer factorization problems, which a cryptanalytically relevant quantum computer could solve efficiently. The US National Institute of Standards and Technology (NIST) has been working on standardizing quantum-resistant algorithms and has selected ML-Key Encapsulation Mechanism (KEM) for key exchange. NIST has also selected ML-DSA for signatures, or cryptographic authentication.
As these algorithms have kilobyte-size public keys, ciphertexts, and signatures — versus the 50- to 400-byte sizes of the existing algorithms — they would inflate the amount of data exchanged in a TLS handshake. A number of works have compared handshake time using traditional TLS 1.3 key exchange and authentication to that using post-quantum (PQ) key exchange and authentication.
These comparisons were useful to quantify the overhead that each new algorithm introduces to the time to first byte, or completion of the handshake protocol. But they ignored the data transfer time over the secure connection that, together with the handshake time, constitutes the total delay before the application can start processing data. The total time from the start of the connection to the end of data transfer is, by contrast, the time to last byte (TTLB). How much TTLB slowdown is acceptable depends highly on the application.
Experiments
We designed our experiments to simulate various network conditions and measured the TTLB of classical and post-quantum algorithms in TLS 1.3 connections where the client makes a small request and the server responds with hundreds of kilobytes (KB) of data. We used Linux namespaces in a Ubuntu 22.04 virtual-machine instance. The namespaces were interconnected using virtual ethernet interfaces. To emulate the “network” between the namespaces, we used the Linux kernel’s netem utility, which can introduce variable network delays, bandwidth fluctuations, and packet loss between the client and server.
Our experiments had several configurable parameters that allowed us to compare the effect of the PQ algorithm on TTLB under stable, unstable, fast, and slow network conditions:
- TLS key exchange mechanism (classical ECDH or ECDH+ML-KEM post-quantum hybrid)
- TLS certificate chain size corresponding to classical RSA or ML-DSA certificates.
- TCP initial congestion window (initcwnd)
- Network delay between client and server, or round-trip time (RTT)
- Bandwidth between client and server
- Loss probability per packet
- Amount of data transferred from the server to the client
Results
The results of our testing are thoroughly analyzed in the paper. They essentially show that a few extra KB in the TLS 1.3 handshake due to the post-quantum public keys, ciphertexts, and signatures will not be noticeable in connections transferring hundreds of KB or more. Connections that transfer less than 10-20 KB of data will probably be more affected by the new data-heavy handshakes.
A bar graph whose y-axis is “handshake time % increase” and whose x-axis is a sequence of percentiles (50th, 75th, and 90th). At each percentile are two bars, one blue (for the traditional handshake protocol) and one orange (for post-quantum handshakes). In all three instances, the orange bar is around twice as high as the blue one.
Figure 1 shows the percentage increase in the duration of the TLS 1.3 handshake for the 50th, 75th, and 90th percentiles of the aggregate datasets collected for 1Mbps bandwidth; 0%, 1%, 3%, and 10% loss probability; and 35-millisecond and 200-millisecond RTT. We can see that the ML-DSA size (16KB) certificate chain takes almost twice as much time as the 8KB chain. This means that if we manage to keep the volume of ML-DSA authentication data low, it would significantly benefit the speed of post-quantum handshakes in low-bandwidth connections.
Figure 2 shows the percentage increase in the duration of the post-quantum handshake relative to the existing algorithm for all percentiles and different data sizes at 0% loss and 1Gbps bandwidth. We can observe that although the slowdown is low (∼3%) at 0 kibibytes (KiB, or multiples of 1,024 bytes, the nearest power of 2 to 1,000) from the server (equivalent to the handshake), it drops even more (∼1%) as the data from the server increases. At the 90th percentile the slowdown is slightly lower.
Figure 3 shows the percentage increase in the TTLB between existing and post-quantum TLS 1.3 connections carrying 0-200KiB of data from the server for each percentile at 1Mbps bandwidth, 200ms RTT, and 0% loss probability. We can see that increases for the three percentiles are almost identical. They start high (∼33%) at 0KiB from the server, but as the data size from the server increases, they drop to ∼6% because the handshake data size is amortized over the connection.
Figure 4 shows the percentage increase in TTLB between existing and post-quantum TLS 1.3 connections carrying 0-200 KiB of data from the server for each percentile at 1Mbps bandwidth, 200ms RTT, and 10% loss probability. It shows that at 10% loss, the TTLB increase settles between 20-30% for all percentiles. The same experiments for 35ms RTT produced similar results. Although a 20-30% increase may seem high, we note that re-running the experiments could sometimes lead to smaller or higher percentage increases because of the general network instability of the scenario. Also, bear in mind that TTLBs for the existing algorithm at 200KiB from the server, 200ms RTT, and 10% loss were 4,644ms, 7,093ms, and 10,178ms, whereas their post-quantum-connection equivalents were 6,010ms, 8,883ms, and 12,378ms. At 0% loss they were 2,364ms, 2,364ms, and 2,364ms. So, although the TTLBs for the post-quantum connections increased by 20-30% relative to the conventional connections, the conventional connections are already impaired (by 97-331%) due to network loss. An extra 20-30% is not likely to make much difference in an already highly degraded connection time.
Figure 5 shows the percentage increase in TTLB between existing and post-quantum TLS 1.3 connections for 0% loss probability and 0-200KiB data sizes transferred from the server. To model a highly volatile RTT, we used a Pareto-normal distribution with a mean of 35ms and 35/4ms jitter. We can see that the increase in post-quantum connection TTLB starts high at 0KiB server data and drops to 4-5%. As with previous experiments, the percentages were more volatile the higher the loss probabilities, but overall, the results show that even under “volatile network conditions” the TTLB drops to acceptable levels as the amount of transferred data increases.
To confirm the volatility under unstable network conditions, we used the TTLB cumulative distribution function (CDF) for post-quantum TLS 1.3 connections transferring 200KiB from the server (figure 6). We observe that under all types of volatile conditions (1Gbps and 5% loss, 1Mbps and 10% loss, Pareto-normal distributed network delay), the TTLB increases very early in the experimental measurement sample, which demonstrates that the total connection times are highly volatile. We made the same observation with TLS 1.3 handshake times under unstable network conditions.
Conclusion
This work demonstrated that the practical effect of data-heavy, post-quantum algorithms on TLS 1.3 connections is lower than their effect on the handshake itself. Low-loss, low- or high-bandwidth connections will see little impact from post-quantum handshakes when transferring sizable amounts of data. We also showed that although the effects of PQ handshakes could vary under unstable conditions with higher loss rates or high-variability delays, they stay within certain limits and drop as the total amount of transferred data increases. Additionally, we saw that unstable connections inherently provide poor completion times; a small latency increase due to post-quantum handshakes would not render them less usable than before. This does not mean that trimming the amount of handshake data is undesirable, especially if little application data is sent relative to the size of the handshake messages.
For more details, please see our paper.
Have you ever worked is legacy code? Are you curious what it takes to modernize systems at a massive scale?
Pascal Hartig is joined on the latest Meta Tech Podcast by Elaine and Buping, two software engineers working on a bold project to rewrite the decades-old C code in one of Meta’s core messaging libraries in Rust. It’s an ambitious effort that will transform a central messaging library that is shared across Messenger, Facebook, Instagram, and Meta’s AR/VR platforms.
They discuss taking on a project of this scope – even without a background in Rust, how they’re approaching it, and what it means to optimize for ‘developer happiness.’
Download or listen to the episode below:
You can also find the episode wherever you get your podcasts, including:
The Meta Tech Podcast is a podcast, brought to you by Meta, where we highlight the work Meta’s engineers are doing at every level – from low-level frameworks to end-user features.
Send us feedback on Instagram, Threads, or X.
And if you’re interested in learning more about career opportunities at Meta visit the Meta Careers page.
Amazon Research Awards (ARA) provides unrestricted funds and AWS Promotional Credits to academic researchers investigating various research topics in multiple disciplines. This cycle, ARA received many excellent research proposals from across the world and today is publicly announcing 73 award recipients who represent 46 universities in 10 countries.
This announcement includes awards funded under five call for proposals during the fall 2024 cycle: AI for Information Security, Automated Reasoning, AWS AI, AWS Cryptography, and Sustainability. Proposals were reviewed for the quality of their scientific content and their potential to impact both the research community and society. Additionally, Amazon encourages the publication of research results, presentations of research at Amazon offices worldwide, and the release of related code under open-source licenses.
Recipients have access to more than 700 Amazon public datasets and can utilize AWS AI/ML services and tools through their AWS Promotional Credits. Recipients also are assigned an Amazon research contact who offers consultation and advice, along with opportunities to participate in Amazon events and training sessions.
“Automated Reasoning is an important area of research for Amazon, with potential applications across various features and applications to help improve security, reliability, and performance for our customers. Through the ARA program, we collaborate with leading academic researchers to explore challenges in this field,” said Robert Jones, senior principal scientist with the Cloud Automated Reasoning Group. “We were again impressed by the exceptional response to our Automated Reasoning call for proposals this year, receiving numerous high-quality submissions. Congratulations to the recipients! We’re excited to support their work and partner with them as they develop new science and technology in this important area.”
“At Amazon, we believe that solving the world’s toughest sustainability challenges benefits from both breakthrough scientific research and open and bold collaboration. Through programs like the Amazon Research Awards program, we aim to support academic research that could contribute to our understanding of these complex issues,” said Kommy Weldemariam, Director of Science and Innovation Sustainability. “The selected proposals represent innovative projects that we hope will help advance knowledge in this field, potentially benefiting customers, communities, and the environment.”
ARA funds proposals throughout the year in a variety of research areas. Applicants are encouraged to visit the ARA call for proposals page for more information or send an email to be notified of future open calls.
The tables below list, in alphabetical order by last name, fall 2024 cycle call-for-proposal recipients, sorted by research area.
AI for Information Security
| Recipient | University | Research title |
| Christopher Amato | Northeastern University | Multi-Agent Reinforcement Learning Cyber Defense for Securing Cloud Computing Platforms |
| Bernd Bischl | Ludwig Maximilian University of Munich | Improving Generative and Foundation Models Reliability via Uncertainty-awareness |
| Shiqing Ma | University Of Massachusetts Amherst | LLM and Domain Adaptation for Attack Detection |
| Alina Oprea | Northeastern University | Multi-Agent Reinforcement Learning Cyber Defense for Securing Cloud Computing Platforms |
| Roberto Perdisci | University of Georgia | ContextADBench: A Comprehensive Benchmark Suite for Contextual Anomaly Detection |
Automated Reasoning
| Recipient | University | Research title |
| Nada Amin | Harvard University | LLM-Augmented Semi-Automated Proofs for Interactive Verification |
| Suguman Bansal | Georgia Institute of Technology | Certified Inductive Generalization in Reinforcement Learning |
| Ioana Boureanu | University of Surrey | Phoebe+: An Automated-Reasoning Tool for Provable Privacy in Cryptographic Systems |
| Omar Haider Chowdhury | Stony Brook University | Restricter: An Automatic Tool for Authoring Amazon Cedar Access Control Policies with the Principle of Least Privilege |
| Stefan Ciobaca | Alexandru Ioan Cuza University | An Interactive Proof Mode for Dafny |
| João Ferreira | INESC-ID | Polyglot Automated Program Repair for Infrastructure as Code |
| Sicun Gao | University Of California, San Diego | Monte Carlo Trees with Conflict Models for Proof Search |
| Mirco Giacobbe | University of Birmingham | Neural Software Verification |
| Tobias Grosser | University of Cambridge | Synthesis-based Symbolic BitVector Simplification for Lean |
| Ronghui Gu | Columbia University | Scaling Formal Verification of Security Properties for Unmodified System Software |
| Alexey Ignatiev | Monash University | Huub: Next-Gen Lazy Clause Generation |
| Kenneth McMillan | University of Texas At Austin | Synthesis of Auxiliary Variables and Invariants for Distributed Protocol Verification |
| Alexandra Mendes | University of Porto | Overcoming Barriers to the Adoption of Verification-Aware Languages |
| Jason Nieh | Columbia University | Scaling Formal Verification of Security Properties for Unmodified System Software |
| Rohan Padhye | Carnegie Mellon University | Automated Synthesis and Evaluation of Property-Based Tests |
| Nadia Polikarpova | University Of California, San Diego | Discovering and Proving Critical System Properties with LLMs |
| Fortunat Rajaona | University of Surrey | Phoebe+: An Automated-Reasoning Tool for Provable Privacy in Cryptographic Systems |
| Subhajit Roy | Indian Institute of Technology Kanpur | Theorem Proving Modulo LLM |
| Gagandeep Singh | University of Illinois At Urbana–Champaign | Trustworthy LLM Systems using Formal Contracts |
| Scott Stoller | Stony Brook University | Restricter: An Automatic Tool for Authoring Amazon Cedar Access Control Policies with the Principle of Least Privilege |
| Peter Stuckey | Monash University | Huub: Next-Gen Lazy Clause Generation |
| Yulei Sui | University of New South Wales | Path-Sensitive Typestate Analysis through Sparse Abstract Execution |
| Nikos Vasilakis | Brown University | Semantics-Driven Static Analysis for the Unix/Linux Shell |
| Ping Wang | Stevens Institute of Technology | Leveraging Large Language Models for Reasoning Augmented Searching on Domain-specific NoSQL Database |
| John Wawrzynek | University of California, Berkeley | GPU-Accelerated High-Throughput SAT Sampling |
AWS AI
| Recipient | University | Research title |
| Panagiotis Adamopoulos | Emory University | Generative AI solutions for The Spillover Effect of Fraudulent Reviews on Product Recommendations |
| Vikram Adve | University of Illinois at Urbana–Champaign | Fellini: Differentiable ML Compiler for Full-Graph Optimization for LLM Models |
| Frances Arnold | California Institute of Technology | Closed-loop Generative Machine Learning for De Novo Enzyme Discovery and Optimization |
| Yonatan Bisk | Carnegie Mellon University | Useful, Safe, and Robust Multiturn Interactions with LLMs |
| Shiyu Chang | University of California, Santa Barbara | Cut the Crap: Advancing the Efficient Communication of Multi-Agent Systems via Spatial-Temporal Topology Design and KV Cache Sharing |
| Yuxin Chen | University of Pennsylvania | Provable Acceleration of Diffusion Models for Modern Generative AI |
| Tianlong Chen | University of North Carolina at Chapel Hill | Cut the Crap: Advancing the Efficient Communication of Multi-Agent Systems via Spatial-Temporal Topology Design and KV Cache Sharing |
| Mingyu Ding | University of North Carolina at Chapel Hill | Aligning Long Videos and Language as Long-Horizon World Models |
| Nikhil Garg | Cornell University | Market Design for Responsible Multi-agent LLMs |
| Jessica Hullman | Northwestern University | Human-Aligned Uncertainty Quantification in High Dimensions |
| Christopher Jermaine | Rice University | Fast, Trusted AI Using the EINSUMMABLE Compiler |
| Yunzhu Li | Columbia University | Physics-Informed Foundation Models Through Embodied Interactions |
| Pattie Maes | Massachusetts Institute of Technology | Understanding How LLM Agents Deviate from Human Choices |
| Sasa Misailovic | University of Illinois at Urbana–Champaign | Fellini: Differentiable ML Compiler for Full-Graph Optimization for LLM Models |
| Kristina Monakhova | Cornell University | Trustworthy extreme imaging for science using interpretable uncertainty quantification |
| Todd Mowry | Carnegie Mellon University | Efficient LLM Serving on Trainium via Kernel Generation |
| Min-hwan Oh | Seoul National University | Mutually Beneficial Interplay Between Selection Fairness and Context Diversity in Contextual Bandits |
| Patrick Rebeschini | University of Oxford | Optimal Regularization for LLM Alignment |
| Jose Renau | University of California, Santa Cruz | Verification Constrained Hardware Optimization using Intelligent Design Agentic Programming |
| Vilma Todri | Emory University | Generative AI solutions for The Spillover Effect of Fraudulent Reviews on Product Recommendations |
| Aravindan Vijayaraghavan | Northwestern University | Human-Aligned Uncertainty Quantification in High Dimensions |
| Wei Yang | University of Texas at Dallas | Optimizing RISC-V Compilers with RISC-LLM and Syntax Parsing |
| Huaxiu Yao | University of North Carolina at Chapel Hill | Aligning Long Videos and Language as Long-Horizon World Models |
| Amy Zhang | University of Washington | Tools for Governing AI Agent Autonomy |
| Ruqi Zhang | Purdue University | Efficient Test-time Alignment for Large Language Models and Large Multimodal Models |
| Zheng Zhang | Rutgers University-New Brunswick | AlphaQC: An AI-powered Quantum Circuit Optimizer and Denoiser |
AWS Cryptography
| Recipient | University | Research title |
| Alexandra Boldyreva | Georgia Institute of Technology | Quantifying Information Leakage in Searchable Encryption Protocols |
| Maria Eichlseder | Graz University of Technology, Austria | SALAD – Systematic Analysis of Lightweight Ascon-based Designs |
| Venkatesan Guruswami | University of California, Berkeley | Obfuscation, Proof Systems, and Secure Computation: A Research Program on Cryptography at the Simons Institute for the Theory of Computing |
| Joseph Jaeger | Georgia Institute of Technology | Analyzing Chat Encryption for Group Messaging |
| Aayush Jain | Carnegie Mellon | Large Scale Multiparty Silent Preprocessing for MPC from LPN |
| Huijia Lin | University of Washington | Large Scale Multiparty Silent Preprocessing for MPC from LPN |
| Hamed Nemati | KTH Royal Institute of Technology | Trustworthy Automatic Verification of Side-Channel Countermeasures for Binary Cryptographic Programs using the HoIBA libary |
| Karl Palmskog | KTH Royal Institute of Technology | Trustworthy Automatic Verification of Side-Channel Countermeasures for Binary Cryptographic Programs using the HoIBA libary |
| Chris Peikert | University of Michigan, Ann Arbor | Practical Third-Generation FHE and Bootstrapping |
| Dimitrios Skarlatos | Carnegie Mellon University | Scale-Out FHE LLMs on GPUs |
| Vinod Vaikuntanathan | Massachusetts Institute of Technology | Can Quantum Computers (Really) Factor? |
| Daniel Wichs | Northeastern University | Obfuscation, Proof Systems, and Secure Computation: A Research Program on Cryptography at the Simons Institute for the Theory of Computing |
| David Wu | University Of Texas At Austin | Fast Private Information Retrieval and More using Homomorphic Encryption |
Sustainability
| Recipient | University | Research title |
| Meeyoung Cha | Max Planck Institute | Forest-Blossom (Flossom): A New Framework for Sustaining Forest Biodiversity Through Outcome-Driven Remote Sensing Monitoring |
| Jingrui He | University of Illinois at Urbana–Champaign | Foundation Model Enabled Earth’s Ecosystem Monitoring |
| Pedro Lopes | University of Chicago | AI-powered Tools that Enable Engineers to Make & Re-make Sustainable Hardware |
| Cheng Yaw Low | Max Planck Institute | Forest-Blossom (Flossom): A New Framework for Sustaining Forest Biodiversity Through Outcome-Driven Remote Sensing Monitoring |
AI safety is a priority at Amazon. Our investment in safe, transparent, and responsible AI (RAI) includes collaboration with the global community and policymakers. We are members of and collaborate with organizations such as the Frontier Model Forum, the Partnership on AI, and other forums organized by government agencies such as the National Institute of Standards and Technology (NIST). Consistent with Amazon’s endorsement of the Korea Frontier AI Safety Commitments, we published our Frontier Model Safety Framework earlier this year.
During the development of the Nova Premier model, we conducted a comprehensive evaluation to assess its performance and safety. This included testing on both internal and public benchmarks and internal/automated and third-party red-teaming exercises. Once the final model was ready, we prioritized obtaining unbiased, third-party evaluations of the model’s robustness against RAI controls. In this post, we outline the key findings from these evaluations, demonstrating the strength of our testing approach and Amazon Premier’s standing as a safe model. Specifically, we cover our evaluations with two third-party evaluators: PRISM AI and ActiveFence.
Evaluation of Nova Premier against PRISM AI
PRISM Eval’s Behavior Elicitation Tool (BET) dynamically and systematically stress-tests AI models’ safety guardrails. The methodology focuses on measuring how many adversarial attempts (steps) it takes to get a model to generate harmful content across several key risk dimensions. The central metric is “steps to elicit” — the number of increasingly sophisticated prompting attempts required before a model generates an inappropriate response. A higher number of steps indicates stronger safety measures, as the model is more resistant to manipulation. The PRISM risk dimensions (inspired by the MLCommons AI Safety Benchmarks) include CBRNE weapons, violent crimes, non-violent crimes, defamation, and hate, amongst several others.
Using the BET Eval tool and its V1.0 metric, which is tailored toward non-reasoning models, we compared the recently released Nova models (Pro and Premier) to the latest models in the same class: Claude (3.5 v2 and 3.7 non-reasoning) and Llama4 Maverick, all available through Amazon Bedrock. PRISM BET conducts black-box evaluations (where model developers don’t have access to the test prompts) of models integrated with their API. The evaluation conducted with BET Eval MAX, PRISM’s most comprehensive/aggressive testing suite, revealed significant variations in safety against malicious instructions. Nova models demonstrated superior overall safety performance, with an average of 43 steps for Premier and 52 steps for Pro, compared to 37.7 for Claude 3.5 v2 and fewer than 12 steps for other models in the comparison set (namely, 9.9 for Claude3.7, 11.5 for Claude 3.7 thinking, and 6.5 for Maverick). This higher step count suggests that on average, Nova’s safety guardrails are more sophisticated and harder to circumvent through adversarial prompting. The figure below presents the number of steps per harm category evaluated through BET Eval MAX.
The PRISM evaluation provides valuable insights into the relative safety of different Amazon Bedrock models. Nova’s strong performance, particularly in hate speech and defamation resistance, represents meaningful progress in AI safety. However, the results also highlight the ongoing challenge of building truly robust safety measures into AI systems. As the field continues to evolve, frameworks like BET will play an increasingly important role in benchmarking and improving AI safety. As a part of this collaboration Nicolas Miailhe, CEO of PRISM Eval, said, “It’s incredibly rewarding for us to see Nova outperforming strong baselines using the BET Eval MAX; our aim is to build a long-term partnership toward safer-by-design models and to make BET available to various model providers.” Organizations deploying AI systems should carefully consider these safety metrics when selecting models for their applications.
Manual red teaming with ActiveFence
The AI safety & security company ActiveFence benchmarked Nova Premier on Bedrock on prompts distributed across Amazon’s eight core RAI categories. ActiveFence also evaluated Claude 3.7 (non-reasoning mode) and GPT 4.1 API on the same set. The flag rate on Nova Premier was lower than that on the other two models, indicating that Nova Premier is the safest of the three.
| Model | 3P Flag Rate [↓ is better] |
| Nova Premier | 12.0% |
| Sonnet 3.7 (non-reasoning) | 20.6% |
| GPT4.1 API | 22.4% |
“Our role is to think like an adversary but act in service of safety,” said Guy Paltieli from ActiveFence. “By conducting a blind stress test of Nova Premier under realistic threat scenarios, we helped evaluate its security posture in support of Amazon’s broader responsible-AI goals, ensuring the model could be deployed with greater confidence.”
These evaluations conducted with PRISM and ActiveFence give us confidence in the strength of our guardrails and our ability to protect our customers’ safety when they use our models. While these evaluations demonstrate strong safety performance, we recognize that AI safety is an ongoing challenge requiring continuous improvement. These assessments represent a point-in-time snapshot, and we remain committed to regular testing and enhancement of our safety measures. No AI system can guarantee perfect safety in all scenarios, which is why we maintain monitoring and response systems after deployment.
Acknowledgments: Vincent Ponzo, Elyssa Vincent
-
Funding & Business6 days agoKayak and Expedia race to build AI travel agents that turn social posts into itineraries
-
Jobs & Careers6 days ago
Mumbai-based Perplexity Alternative Has 60k+ Users Without Funding
-
Mergers & Acquisitions6 days ago
Donald Trump suggests US government review subsidies to Elon Musk’s companies
-
Funding & Business6 days ago
Rethinking Venture Capital’s Talent Pipeline
-
Jobs & Careers6 days ago
Why Agentic AI Isn’t Pure Hype (And What Skeptics Aren’t Seeing Yet)
-
Funding & Business3 days ago
Sakana AI’s TreeQuest: Deploy multi-model teams that outperform individual LLMs by 30%
-
Funding & Business6 days agoFrom chatbots to collaborators: How AI agents are reshaping enterprise work
-
Funding & Business6 days ago
Europe’s Most Ambitious Startups Aren’t Becoming Global; They’re Starting That Way
-
Tools & Platforms6 days ago
Winning with AI – A Playbook for Pest Control Business Leaders to Drive Growth
-
Jobs & Careers4 days ago
Ilya Sutskever Takes Over as CEO of Safe Superintelligence After Daniel Gross’s Exit
